Dr. Eidolon

21.3K posts

Dr. Eidolon banner
Dr. Eidolon

Dr. Eidolon

@DrEidolon

Civilizations rise through intelligence, energy and technology. The future belongs to adaptive systems.

Katılım Aralık 2021
3.6K Takip Edilen3.8K Takipçiler
Sabitlenmiş Tweet
Dr. Eidolon
Dr. Eidolon@DrEidolon·
$SIVE — Why the Thesis Holds A thread responding to the “Sivers isn’t a chokepoint” take that’s been circulating. ----- The critique rests on three claims: 1. Hyperscalers multi-source, so Sivers can always be routed around. 1. Sivers doesn’t own fabs, so the real chokepoint sits upstream and applies to Sivers too. 1. The IP lead will erode before the window opens. These aren’t entirely wrong. The conclusion drawn from them is. ----- On “chokepoint” vs “monopoly” The critique quietly redefines chokepoint as absolute, permanent irreplaceability. That’s not how supply chains work. A structural chokepoint is a node that is difficult to replace within a competitive timeframe, at the required scale, with qualified alternatives. The global capacity to produce qualified DFB CW laser arrays for CPO architectures is extraordinarily concentrated. That’s not a claim. It’s the reason Lumentum disclosed it had to purchase CW lasers on the open market from competitors — its own capacity was fully allocated to Nvidia. If abundant, easily qualified alternatives existed, Lumentum wouldn’t be buying externally. Multi-sourcing doesn’t contradict this. It confirms it. Hyperscalers multi-source because supply is tight and they need risk coverage. Being in the top-2 qualified laser suppliers across Ayar, Jabil, POET, and Marvell Celestial simultaneously is not “a little bit everywhere, married to no one.” It is a structural position that takes years to build and cannot be replicated overnight. ----- On the fab argument Yes, Sivers doesn’t own its fabs. Neither does Marvell, AMD, or Broadcom. The question is not ownership. It’s allocated capacity. Sivers secured its Win Semiconductor partnership in March 2025 specifically to lock in high-volume production capacity ahead of the CPO ramp. Lumentum, by contrast, is capacity-constrained despite owning fabs — because its internal lines are saturated by Nvidia. In the 2026-2028 window, which is the only window that matters for this thesis, Sivers is better positioned on capacity than the critique implies. ----- On qualification moats The critic argues IP leads erode as larger competitors scale up. Generically true. But it misses the most important barrier in this market: customer qualification. Qualification cycles in CPO laser supply chains run 12 to 24 months minimum. A competitor who closes the IP gap does not automatically inherit the qualification. Sivers is already qualified inside the stacks of Ayar Labs, Jabil, POET, and Marvell Celestial. Each of those customers has a strong incentive not to re-qualify a new supplier mid-ramp. The risk of supply disruption is too high. And per Wayback Machine analysis of Ayar’s supplier page: Lumentum and MACOM were silently removed. Sivers is now the only publicly listed laser supplier to Ayar — which just raised $500M to scale volumes. That is not a footnote. ----- On the confirmation bias argument The critic closes by suggesting that anyone who believes the thesis is a victim of confirmation bias, following a persuasive narrator and losing critical thinking over time. This is a rhetorical move, not an analytical one. It invalidates the argument based on the supposed psychology of those who hold it rather than on the evidence itself. The customer mapping, the Wayback Machine supplier data, the Win Semi capacity allocation, the Jabil 1.6T announcement, the Marvell Celestial direct relationship evidence — these are either factually accurate or they aren’t. Whether people want them to be true is irrelevant to whether they are. The sceptic who assumes the contrarian position is automatically more rational is applying his own confirmation bias in reverse. ----- $SIVE @aleabitoreddit
Moody@MoodyWriter13

That’s a nice overview, and clearly a lot of work went into it. It shows the many relationships Sivers has. But he draws the wrong conclusion from it. He says, “Sivers is the CPO chokepoint for Hyperscalers.” It isn’t. A chokepoint is a place everyone has to pass through and nobody can route around. Sivers is involved a little bit everywhere, but irreplaceable nowhere. Whoever talks to everyone is firmly married to no one. He even writes it himself: the big customers always buy from several suppliers at once. If you’re replaceable, you’re not a chokepoint, and you have little pricing power. As I described in my article, the real chokepoint sits one stage higher, and that’s where Sivers itself is standing in line. There is no shortage of laser designs. There is a shortage of factories to build them. And Sivers doesn’t own those factories. at that stage Sivers is just a customer. The chokepoint belongs to others: the firms with their own fabs (Coherent, Lumentum), the equipment makers behind them (Aixtron), and the substrate suppliers. The fact that Lumentum currently even has to buy lasers on the open market, because its own capacity is being prioritized toward Nvidia, could indeed help Sivers in the short term, but that’s a Phase 1 effect, not a permanent state. You could argue the opposite, that Sivers profits precisely from this. But once you ask why Lumentum has to reach into the market in the first place, the answer leads right back to the actual chokepoints, and those apply to Sivers even more than to Lumentum. Sivers’ strong position as one of currently few firms with the relevant laser IP will also weaken over time, because it’s reasonable to expect that more large competitors will close the gap. The real chokepoints further down in production prevent Sivers from exploiting this lead during this phase. And by the time those chokepoints are resolved, the competition will probably already have caught up. The company can still succeed, no question. But the claim that Sivers is a permanently indispensable chokepoint is simply wrong. Serenity probably knows this himself, and is deliberately framing it that way. Everyone who has followed him so far, and who also has a stake in Sivers succeeding, hopes that what he says is true. He’s exploiting confirmation bias. So don’t forget to question things critically. Whoever stops doing that loses the ability over time entirely, and then your wellbeing always hangs on the people you follow. Sivers is well connected, but not irreplaceable. Those are two different things, and on the market that difference is what matters in the long run.

English
12
12
148
30.6K
The Fundamentalist
The Fundamentalist@Funmentalist·
Serenity surprised $SIVE is -15% I will give you a hint OVERVALUED, SPECULATIVE GOD, I HATE THIS GUY He has insane picks, do not get me wrong Insane calls, but no one is above God, not even him He feels like some kind of genius because every pick goes up, guess what? If you have 400k followers and you talk about a $500M market cap stock, it will go up together with you pumping in cash
English
29
1
30
8.1K
Alex
Alex@zGwmby·
@aleabitoreddit most likely money is rotating into $XFAB from $sive?
English
5
0
8
3K
Serenity
Serenity@aleabitoreddit·
Interesting photonics selloff today on no news? $LITE down -4.95% $AAOI down -4.85% $SIVE down -14.8% $SOI down -5.73% $AXTI down -8.13% $IQE down -12.13% I think it’s probably the most compelling theme going forward (even more than power semis). Just tends to be very volatile on the way up. Surprised about $AAOI though given there’s some institutional notes apparently about long term $AMD or $NVDA agreements. (Rosenblatt). Maybe $600m ATM caps some near term upside. $SIVE as well, given EU Chips Act 2 is next week around photonics, and they’re listed on the blueprint. Same with MSCI/NASDAQ omx inflow next week. I’ve been personally adding to positions since I have high conviction in the photonics theme (CPO especially) given TAM expansion overall next 2 years.
Serenity tweet mediaSerenity tweet media
English
285
130
1.8K
295.6K
Serenity
Serenity@aleabitoreddit·
Yeah would not be surprised to see $SIVE, $SOI, and $XFAB funding with EU Chips Act 2. Glad to all my longs there (aside from $IQE) listed on the blueprint, I didn’t see that earlier in my research! Looks like formal announcement got pushed back to next week though.
Serenity tweet media
Vnz@lesantcho9306

@aleabitoreddit It was included in the March 2026 blueprint. Let’s see the publication of today and the proposal for EU CHIPS ACT 2.0

English
122
58
783
358.5K
Neural Space
Neural Space@NeuralSpace_·
Les yeux gris et verts sont parmi les couleurs les plus rares chez les humains. Environ 3 % de la population mondiale a des yeux gris et 2 % ont des yeux verts.
Français
5
11
87
3.4K
Picked By Nik
Picked By Nik@pickedbynik·
I have seen a lot of things in my life, and I thought Tesla cars were gay, but listen, this new Ferrari thing is so gay that even gays look straight. What the actual... (finish the sentence in comments)
Picked By Nik tweet media
English
13
18
92
70.2K
SlarkTrader
SlarkTrader@SlarkTrader·
I have entered a new position for the summer: —> 90% success rate of FDA approval within the next 90 days —> $2.4 BILLION dollar market —> 100-200% short term 90 day target —> 1700% 2 year target Subscribe to join me and follow my plan slarktrader.substack.com
English
11
0
13
7.1K
BillieBillion
BillieBillion@BillieBillion1·
To anyone who forgot I want to remind you of when @aleabitoreddit got illegal insider information from $SIVE management and accidentally told everyone about it on X. Reminds me of when Hans Isoz tweeted about a text message he got from the CEO of Intellego informing him about an upcoming article in @dagensindustri - the similarities are quite striking🤔
BillieBillion tweet media
English
38
1
49
65.5K
AlertesInfos
AlertesInfos@AlertesInfos·
🇫🇷🕊️ FLASH | L’animateur Arthur annonce la mort de son père, Michel Essebag. (Le Parisien)
AlertesInfos tweet media
Français
585
103
2.7K
1.9M
Dr. Eidolon retweetledi
COPE
COPE@COPE·
🏦 El FMI lanza una seria advertencia a España por su política de vivienda 📉 El organismo internacional pide revisar el tope a los alquileres al considerar que reduce la oferta cope.es/actualidad/eco…
Español
20
58
202
258.9K
Dr. Eidolon retweetledi
Steven Arrazola de Oñate
HLN deelde eerder vandaag op Instagram, en vermoedelijk ook op andere platformen, deze beelden van de kusttram in De Haan, maar dan geblurd. De beelden gingen viraal en de commentaren waren logischerwijs niet bepaald vriendelijk voor dit krapuul uit Brussel. HLN heeft de beelden ondertussen van Instagram verwijderd en bij de meeste andere kranten kan je niet meer reageren op artikels over deze degoutante gebeurtenis. Omdat ik hier zo kwaad van word, deel ik hier graag de niet-geblurde beelden. Waarom blijven onze politici en de MSM dit probleem ontkennen én minimaliseren???
Nederlands
343
1.3K
4.1K
238.4K
Dr. Eidolon retweetledi
Xander
Xander@BeOstbelgien·
„Plus tôt dans la journée, HLN a partagé ces images du tramway côtier de De Haan sur Instagram, et probablement aussi sur d'autres plateformes, mais floutées. Les images sont devenues virales et, sans surprise, les commentaires ont été loin d'être tendres envers cette racaille bruxelloise. HLN a depuis retiré les images d'Instagram et, sur la plupart des autres journaux, il n'est plus possible de commenter les articles relatant cet incident révoltant. Furieux, je souhaite partager ici les images non floutées. Pourquoi nos politiciens et les grands médias persistent-ils à nier et minimiser ce problème ?“
Steven Arrazola de Oñate@StevenArra

HLN deelde eerder vandaag op Instagram, en vermoedelijk ook op andere platformen, deze beelden van de kusttram in De Haan, maar dan geblurd. De beelden gingen viraal en de commentaren waren logischerwijs niet bepaald vriendelijk voor dit krapuul uit Brussel. HLN heeft de beelden ondertussen van Instagram verwijderd en bij de meeste andere kranten kan je niet meer reageren op artikels over deze degoutante gebeurtenis. Omdat ik hier zo kwaad van word, deel ik hier graag de niet-geblurde beelden. Waarom blijven onze politici en de MSM dit probleem ontkennen én minimaliseren???

Français
8
30
93
2.2K
Dr. Eidolon retweetledi
System 🤘
System 🤘@SystemLean·
L’IA est en train de faire ce que les planificateurs centraux ont toujours promis sans jamais y arriver : rendre l’abondance structurelle. Et elle le fait en contournant précisément le système qu’ils contrôlent. C’est pas de la désinflation. C’est une révolution silencieuse du paradigme de rareté.
Chubby♨️@kimmonismus

„AI is going to make everything cost less.” The new Fed Chair says the next few years could be a hugely exciting moment, arguing that AI is “structurally disinflationary.” Read that again: „structurally disinflationary“. And that from the new Fed Chair.

Français
6
8
23
2K